Table 10.

AI outcomes in endodontics.

Target	AI Model	Sample	Results	Study
Locating the minor apical foramen	ANN	50 teeth	To enhance the accuracy of working length measurement using radiography, artificial neural networks can serve as a second opinion to find the apical foramen on radiographs.	Saghiri et al., (2012) [199]
Vertical root fracture detection	ANN	Digital X-rays: 50 sound and 150 vertical root fractures	Adequate sensitivity, specificity, and accuracy to be used as a model for vertical root fracture detection.	Kositbowornchai et al., (2013) [200]
Detecting vertical root fracture on X-ray images of endodontically treated and intact teeth	PNN	240 radiographs (120/120)	96.6% accuracy, 93.3% sensitivity, 100% specificity.	Johari et al., (2017) [198]
Detecting vertical root fracture on panoramic radiography	CNN	300 panoramic images	Precision of 0.93	Fukuda et al., (2020) [201]
To detect periapical pathosis	CNN	153 CBCT images	Accuracy of 92.8%	Orhan et al., (2020) [202]

AI, artificial intelligence; ANN, artificial neural network; CBCT, cone-beam computed tomography; CI, confidence interval; CNN, convolutional neural network; PNN, probabilistic neural network.